includes/content/ContentHandler.php

   1 <?php
   2 /**
   3  * Base class for content handling.
   4  *
   5  * This program is free software; you can redistribute it and/or modify
   6  * it under the terms of the GNU General Public License as published by
   7  * the Free Software Foundation; either version 2 of the License, or
   8  * (at your option) any later version.
   9  *
  10  * This program is distributed in the hope that it will be useful,
  11  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  12  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  13  * GNU General Public License for more details.
  14  *
  15  * You should have received a copy of the GNU General Public License along
  16  * with this program; if not, write to the Free Software Foundation, Inc.,
  17  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  18  * http://www.gnu.org/copyleft/gpl.html
  19  *
  20  * @since 1.21
  21  *
  22  * @file
  23  * @ingroup Content
  24  *
  25  * @author Daniel Kinzler
  26  */
  27
  28 use Wikimedia\Assert\Assert;
  29 use MediaWiki\Logger\LoggerFactory;
  30 use MediaWiki\MediaWikiServices;
  31 use MediaWiki\Revision\SlotRenderingProvider;
  32 use MediaWiki\Search\ParserOutputSearchDataExtractor;
  33
  34 /**
  35  * A content handler knows how do deal with a specific type of content on a wiki
  36  * page. Content is stored in the database in a serialized form (using a
  37  * serialization format a.k.a. MIME type) and is unserialized into its native
  38  * PHP representation (the content model), which is wrapped in an instance of
  39  * the appropriate subclass of Content.
  40  *
  41  * ContentHandler instances are stateless singletons that serve, among other
  42  * things, as a factory for Content objects. Generally, there is one subclass
  43  * of ContentHandler and one subclass of Content for every type of content model.
  44  *
  45  * Some content types have a flat model, that is, their native representation
  46  * is the same as their serialized form. Examples would be JavaScript and CSS
  47  * code. As of now, this also applies to wikitext (MediaWiki's default content
  48  * type), but wikitext content may be represented by a DOM or AST structure in
  49  * the future.
  50  *
  51  * @ingroup Content
  52  */
  53 abstract class ContentHandler {
  54         /**
  55          * Convenience function for getting flat text from a Content object. This
  56          * should only be used in the context of backwards compatibility with code
  57          * that is not yet able to handle Content objects!
  58          *
  59          * If $content is null, this method returns the empty string.
  60          *
  61          * If $content is an instance of TextContent, this method returns the flat
  62          * text as returned by $content->getText().
  63          *
  64          * If $content is not a TextContent object, the behavior of this method
  65          * depends on the global $wgContentHandlerTextFallback:
  66          * - If $wgContentHandlerTextFallback is 'fail' and $content is not a
  67          *   TextContent object, an MWException is thrown.
  68          * - If $wgContentHandlerTextFallback is 'serialize' and $content is not a
  69          *   TextContent object, $content->serialize() is called to get a string
  70          *   form of the content.
  71          * - If $wgContentHandlerTextFallback is 'ignore' and $content is not a
  72          *   TextContent object, this method returns null.
  73          * - otherwise, the behavior is undefined.
  74          *
  75          * @since 1.21
  76          *
  77          * @param Content|null $content
  78          *
  79          * @throws MWException If the content is not an instance of TextContent and
  80          * wgContentHandlerTextFallback was set to 'fail'.
  81          * @return string|null Textual form of the content, if available.
  82          */
  83         public static function getContentText( Content $content = null ) {
  84                 global $wgContentHandlerTextFallback;
  85
  86                 if ( is_null( $content ) ) {
  87                         return '';
  88                 }
  89
  90                 if ( $content instanceof TextContent ) {
  91                         return $content->getText();
  92                 }
  93
  94                 wfDebugLog( 'ContentHandler', 'Accessing ' . $content->getModel() . ' content as text!' );
  95
  96                 if ( $wgContentHandlerTextFallback == 'fail' ) {
  97                         throw new MWException(
  98                                 "Attempt to get text from Content with model " .
  99                                 $content->getModel()
 100                         );
 101                 }
 102
 103                 if ( $wgContentHandlerTextFallback == 'serialize' ) {
 104                         return $content->serialize();
 105                 }
 106
 107                 return null;
 108         }
 109
 110         /**
 111          * Convenience function for creating a Content object from a given textual
 112          * representation.
 113          *
 114          * $text will be deserialized into a Content object of the model specified
 115          * by $modelId (or, if that is not given, $title->getContentModel()) using
 116          * the given format.
 117          *
 118          * @since 1.21
 119          *
 120          * @param string $text The textual representation, will be
 121          *    unserialized to create the Content object
 122          * @param Title|null $title The title of the page this text belongs to.
 123          *    Required if $modelId is not provided.
 124          * @param string|null $modelId The model to deserialize to. If not provided,
 125          *    $title->getContentModel() is used.
 126          * @param string|null $format The format to use for deserialization. If not
 127          *    given, the model's default format is used.
 128          *
 129          * @throws MWException If model ID or format is not supported or if the text can not be
 130          * unserialized using the format.
 131          * @return Content A Content object representing the text.
 132          */
 133         public static function makeContent( $text, Title $title = null,
 134                 $modelId = null, $format = null ) {
 135                 if ( is_null( $modelId ) ) {
 136                         if ( is_null( $title ) ) {
 137                                 throw new MWException( "Must provide a Title object or a content model ID." );
 138                         }
 139
 140                         $modelId = $title->getContentModel();
 141                 }
 142
 143                 $handler = self::getForModelID( $modelId );
 144
 145                 return $handler->unserializeContent( $text, $format );
 146         }
 147
 148         /**
 149          * Returns the name of the default content model to be used for the page
 150          * with the given title.
 151          *
 152          * Note: There should rarely be need to call this method directly.
 153          * To determine the actual content model for a given page, use
 154          * Title::getContentModel().
 155          *
 156          * Which model is to be used by default for the page is determined based
 157          * on several factors:
 158          * - The global setting $wgNamespaceContentModels specifies a content model
 159          *   per namespace.
 160          * - The hook ContentHandlerDefaultModelFor may be used to override the page's default
 161          *   model.
 162          * - Pages in NS_MEDIAWIKI and NS_USER default to the CSS or JavaScript
 163          *   model if they end in .js or .css, respectively.
 164          * - Pages in NS_MEDIAWIKI default to the wikitext model otherwise.
 165          * - The hook TitleIsCssOrJsPage may be used to force a page to use the CSS
 166          *   or JavaScript model. This is a compatibility feature. The ContentHandlerDefaultModelFor
 167          *   hook should be used instead if possible.
 168          * - The hook TitleIsWikitextPage may be used to force a page to use the
 169          *   wikitext model. This is a compatibility feature. The ContentHandlerDefaultModelFor
 170          *   hook should be used instead if possible.
 171          *
 172          * If none of the above applies, the wikitext model is used.
 173          *
 174          * Note: this is used by, and may thus not use, Title::getContentModel()
 175          *
 176          * @since 1.21
 177          * @deprecated since 1.33, use SlotRoleHandler::getDefaultModel() together with
 178          * SlotRoleRegistry::getRoleHandler().
 179          *
 180          * @param Title $title
 181          *
 182          * @return string Default model name for the page given by $title
 183          */
 184         public static function getDefaultModelFor( Title $title ) {
 185                 $slotRoleregistry = MediaWikiServices::getInstance()->getSlotRoleRegistry();
 186                 $mainSlotHandler = $slotRoleregistry->getRoleHandler( 'main' );
 187                 return $mainSlotHandler->getDefaultModel( $title );
 188         }
 189
 190         /**
 191          * Returns the appropriate ContentHandler singleton for the given title.
 192          *
 193          * @since 1.21
 194          *
 195          * @param Title $title
 196          *
 197          * @return ContentHandler
 198          */
 199         public static function getForTitle( Title $title ) {
 200                 $modelId = $title->getContentModel();
 201
 202                 return self::getForModelID( $modelId );
 203         }
 204
 205         /**
 206          * Returns the appropriate ContentHandler singleton for the given Content
 207          * object.
 208          *
 209          * @since 1.21
 210          *
 211          * @param Content $content
 212          *
 213          * @return ContentHandler
 214          */
 215         public static function getForContent( Content $content ) {
 216                 $modelId = $content->getModel();
 217
 218                 return self::getForModelID( $modelId );
 219         }
 220
 221         /**
 222          * @var array A Cache of ContentHandler instances by model id
 223          */
 224         protected static $handlers;
 225
 226         /**
 227          * Returns the ContentHandler singleton for the given model ID. Use the
 228          * CONTENT_MODEL_XXX constants to identify the desired content model.
 229          *
 230          * ContentHandler singletons are taken from the global $wgContentHandlers
 231          * array. Keys in that array are model names, the values are either
 232          * ContentHandler singleton objects, or strings specifying the appropriate
 233          * subclass of ContentHandler.
 234          *
 235          * If a class name is encountered when looking up the singleton for a given
 236          * model name, the class is instantiated and the class name is replaced by
 237          * the resulting singleton in $wgContentHandlers.
 238          *
 239          * If no ContentHandler is defined for the desired $modelId, the
 240          * ContentHandler may be provided by the ContentHandlerForModelID hook.
 241          * If no ContentHandler can be determined, an MWException is raised.
 242          *
 243          * @since 1.21
 244          *
 245          * @param string $modelId The ID of the content model for which to get a
 246          *    handler. Use CONTENT_MODEL_XXX constants.
 247          *
 248          * @throws MWException For internal errors and problems in the configuration.
 249          * @throws MWUnknownContentModelException If no handler is known for the model ID.
 250          * @return ContentHandler The ContentHandler singleton for handling the model given by the ID.
 251          */
 252         public static function getForModelID( $modelId ) {
 253                 global $wgContentHandlers;
 254
 255                 if ( isset( self::$handlers[$modelId] ) ) {
 256                         return self::$handlers[$modelId];
 257                 }
 258
 259                 if ( empty( $wgContentHandlers[$modelId] ) ) {
 260                         $handler = null;
 261
 262                         Hooks::run( 'ContentHandlerForModelID', [ $modelId, &$handler ] );
 263
 264                         if ( $handler === null ) {
 265                                 throw new MWUnknownContentModelException( $modelId );
 266                         }
 267
 268                         if ( !( $handler instanceof ContentHandler ) ) {
 269                                 throw new MWException( "ContentHandlerForModelID must supply a ContentHandler instance" );
 270                         }
 271                 } else {
 272                         $classOrCallback = $wgContentHandlers[$modelId];
 273
 274                         if ( is_callable( $classOrCallback ) ) {
 275                                 $handler = call_user_func( $classOrCallback, $modelId );
 276                         } else {
 277                                 $handler = new $classOrCallback( $modelId );
 278                         }
 279
 280                         if ( !( $handler instanceof ContentHandler ) ) {
 281                                 throw new MWException( "$classOrCallback from \$wgContentHandlers is not " .
 282                                         "compatible with ContentHandler" );
 283                         }
 284                 }
 285
 286                 wfDebugLog( 'ContentHandler', 'Created handler for ' . $modelId
 287                         . ': ' . get_class( $handler ) );
 288
 289                 self::$handlers[$modelId] = $handler;
 290
 291                 return self::$handlers[$modelId];
 292         }
 293
 294         /**
 295          * Clean up handlers cache.
 296          */
 297         public static function cleanupHandlersCache() {
 298                 self::$handlers = [];
 299         }
 300
 301         /**
 302          * Returns the localized name for a given content model.
 303          *
 304          * Model names are localized using system messages. Message keys
 305          * have the form content-model-$name, where $name is getContentModelName( $id ).
 306          *
 307          * @param string $name The content model ID, as given by a CONTENT_MODEL_XXX
 308          *    constant or returned by Revision::getContentModel().
 309          * @param Language|null $lang The language to parse the message in (since 1.26)
 310          *
 311          * @throws MWException If the model ID isn't known.
 312          * @return string The content model's localized name.
 313          */
 314         public static function getLocalizedName( $name, Language $lang = null ) {
 315                 // Messages: content-model-wikitext, content-model-text,
 316                 // content-model-javascript, content-model-css
 317                 $key = "content-model-$name";
 318
 319                 $msg = wfMessage( $key );
 320                 if ( $lang ) {
 321                         $msg->inLanguage( $lang );
 322                 }
 323
 324                 return $msg->exists() ? $msg->plain() : $name;
 325         }
 326
 327         public static function getContentModels() {
 328                 global $wgContentHandlers;
 329
 330                 $models = array_keys( $wgContentHandlers );
 331                 Hooks::run( 'GetContentModels', [ &$models ] );
 332                 return $models;
 333         }
 334
 335         public static function getAllContentFormats() {
 336                 global $wgContentHandlers;
 337
 338                 $formats = [];
 339
 340                 foreach ( $wgContentHandlers as $model => $class ) {
 341                         $handler = self::getForModelID( $model );
 342                         $formats = array_merge( $formats, $handler->getSupportedFormats() );
 343                 }
 344
 345                 $formats = array_unique( $formats );
 346
 347                 return $formats;
 348         }
 349
 350         // ------------------------------------------------------------------------
 351
 352         /**
 353          * @var string
 354          */
 355         protected $mModelID;
 356
 357         /**
 358          * @var string[]
 359          */
 360         protected $mSupportedFormats;
 361
 362         /**
 363          * Constructor, initializing the ContentHandler instance with its model ID
 364          * and a list of supported formats. Values for the parameters are typically
 365          * provided as literals by subclass's constructors.
 366          *
 367          * @param string $modelId (use CONTENT_MODEL_XXX constants).
 368          * @param string[] $formats List for supported serialization formats
 369          *    (typically as MIME types)
 370          */
 371         public function __construct( $modelId, $formats ) {
 372                 $this->mModelID = $modelId;
 373                 $this->mSupportedFormats = $formats;
 374         }
 375
 376         /**
 377          * Serializes a Content object of the type supported by this ContentHandler.
 378          *
 379          * @since 1.21
 380          *
 381          * @param Content $content The Content object to serialize
 382          * @param string|null $format The desired serialization format
 383          *
 384          * @return string Serialized form of the content
 385          */
 386         abstract public function serializeContent( Content $content, $format = null );
 387
 388         /**
 389          * Applies transformations on export (returns the blob unchanged per default).
 390          * Subclasses may override this to perform transformations such as conversion
 391          * of legacy formats or filtering of internal meta-data.
 392          *
 393          * @param string $blob The blob to be exported
 394          * @param string|null $format The blob's serialization format
 395          *
 396          * @return string
 397          */
 398         public function exportTransform( $blob, $format = null ) {
 399                 return $blob;
 400         }
 401
 402         /**
 403          * Unserializes a Content object of the type supported by this ContentHandler.
 404          *
 405          * @since 1.21
 406          *
 407          * @param string $blob Serialized form of the content
 408          * @param string|null $format The format used for serialization
 409          *
 410          * @return Content The Content object created by deserializing $blob
 411          */
 412         abstract public function unserializeContent( $blob, $format = null );
 413
 414         /**
 415          * Apply import transformation (per default, returns $blob unchanged).
 416          * This gives subclasses an opportunity to transform data blobs on import.
 417          *
 418          * @since 1.24
 419          *
 420          * @param string $blob
 421          * @param string|null $format
 422          *
 423          * @return string
 424          */
 425         public function importTransform( $blob, $format = null ) {
 426                 return $blob;
 427         }
 428
 429         /**
 430          * Creates an empty Content object of the type supported by this
 431          * ContentHandler.
 432          *
 433          * @since 1.21
 434          *
 435          * @return Content
 436          */
 437         abstract public function makeEmptyContent();
 438
 439         /**
 440          * Creates a new Content object that acts as a redirect to the given page,
 441          * or null if redirects are not supported by this content model.
 442          *
 443          * This default implementation always returns null. Subclasses supporting redirects
 444          * must override this method.
 445          *
 446          * Note that subclasses that override this method to return a Content object
 447          * should also override supportsRedirects() to return true.
 448          *
 449          * @since 1.21
 450          *
 451          * @param Title $destination The page to redirect to.
 452          * @param string $text Text to include in the redirect, if possible.
 453          *
 454          * @return Content Always null.
 455          */
 456         public function makeRedirectContent( Title $destination, $text = '' ) {
 457                 return null;
 458         }
 459
 460         /**
 461          * Returns the model id that identifies the content model this
 462          * ContentHandler can handle. Use with the CONTENT_MODEL_XXX constants.
 463          *
 464          * @since 1.21
 465          *
 466          * @return string The model ID
 467          */
 468         public function getModelID() {
 469                 return $this->mModelID;
 470         }
 471
 472         /**
 473          * @since 1.21
 474          *
 475          * @param string $model_id The model to check
 476          *
 477          * @throws MWException If the model ID is not the ID of the content model supported by this
 478          * ContentHandler.
 479          */
 480         protected function checkModelID( $model_id ) {
 481                 if ( $model_id !== $this->mModelID ) {
 482                         throw new MWException( "Bad content model: " .
 483                                 "expected {$this->mModelID} " .
 484                                 "but got $model_id." );
 485                 }
 486         }
 487
 488         /**
 489          * Returns a list of serialization formats supported by the
 490          * serializeContent() and unserializeContent() methods of this
 491          * ContentHandler.
 492          *
 493          * @since 1.21
 494          *
 495          * @return string[] List of serialization formats as MIME type like strings
 496          */
 497         public function getSupportedFormats() {
 498                 return $this->mSupportedFormats;
 499         }
 500
 501         /**
 502          * The format used for serialization/deserialization by default by this
 503          * ContentHandler.
 504          *
 505          * This default implementation will return the first element of the array
 506          * of formats that was passed to the constructor.
 507          *
 508          * @since 1.21
 509          *
 510          * @return string The name of the default serialization format as a MIME type
 511          */
 512         public function getDefaultFormat() {
 513                 return $this->mSupportedFormats[0];
 514         }
 515
 516         /**
 517          * Returns true if $format is a serialization format supported by this
 518          * ContentHandler, and false otherwise.
 519          *
 520          * Note that if $format is null, this method always returns true, because
 521          * null means "use the default format".
 522          *
 523          * @since 1.21
 524          *
 525          * @param string $format The serialization format to check
 526          *
 527          * @return bool
 528          */
 529         public function isSupportedFormat( $format ) {
 530                 if ( !$format ) {
 531                         return true; // this means "use the default"
 532                 }
 533
 534                 return in_array( $format, $this->mSupportedFormats );
 535         }
 536
 537         /**
 538          * Convenient for checking whether a format provided as a parameter is actually supported.
 539          *
 540          * @param string $format The serialization format to check
 541          *
 542          * @throws MWException If the format is not supported by this content handler.
 543          */
 544         protected function checkFormat( $format ) {
 545                 if ( !$this->isSupportedFormat( $format ) ) {
 546                         throw new MWException(
 547                                 "Format $format is not supported for content model "
 548                                 . $this->getModelID()
 549                         );
 550                 }
 551         }
 552
 553         /**
 554          * Returns overrides for action handlers.
 555          * Classes listed here will be used instead of the default one when
 556          * (and only when) $wgActions[$action] === true. This allows subclasses
 557          * to override the default action handlers.
 558          *
 559          * @since 1.21
 560          *
 561          * @return array An array mapping action names (typically "view", "edit", "history" etc.) to
 562          *  either the full qualified class name of an Action class, a callable taking ( Page $page,
 563          *  IContextSource $context = null ) as parameters and returning an Action object, or an actual
 564          *  Action object. An empty array in this default implementation.
 565          *
 566          * @see Action::factory
 567          */
 568         public function getActionOverrides() {
 569                 return [];
 570         }
 571
 572         /**
 573          * Factory for creating an appropriate DifferenceEngine for this content model.
 574          * Since 1.32, this is only used for page-level diffs; to diff two content objects,
 575          * use getSlotDiffRenderer.
 576          *
 577          * The DifferenceEngine subclass to use is selected in getDiffEngineClass(). The
 578          * GetDifferenceEngine hook will receive the DifferenceEngine object and can replace or
 579          * wrap it.
 580          * (Note that in older versions of MediaWiki the hook documentation instructed extensions
 581          * to return false from the hook; you should not rely on always being able to decorate
 582          * the DifferenceEngine instance from the hook. If the owner of the content type wants to
 583          * decorare the instance, overriding this method is a safer approach.)
 584          *
 585          * @todo This is page-level functionality so it should not belong to ContentHandler.
 586          *   Move it to a better place once one exists (e.g. PageTypeHandler).
 587          *
 588          * @since 1.21
 589          *
 590          * @param IContextSource $context Context to use, anything else will be ignored.
 591          * @param int $old Revision ID we want to show and diff with.
 592          * @param int|string $new Either a revision ID or one of the strings 'cur', 'prev' or 'next'.
 593          * @param int $rcid FIXME: Deprecated, no longer used. Defaults to 0.
 594          * @param bool $refreshCache If set, refreshes the diff cache. Defaults to false.
 595          * @param bool $unhide If set, allow viewing deleted revs. Defaults to false.
 596          *
 597          * @return DifferenceEngine
 598          */
 599         public function createDifferenceEngine( IContextSource $context, $old = 0, $new = 0,
 600                 $rcid = 0, // FIXME: Deprecated, no longer used
 601                 $refreshCache = false, $unhide = false
 602         ) {
 603                 $diffEngineClass = $this->getDiffEngineClass();
 604                 $differenceEngine = new $diffEngineClass( $context, $old, $new, $rcid, $refreshCache, $unhide );
 605                 Hooks::run( 'GetDifferenceEngine', [ $context, $old, $new, $refreshCache, $unhide,
 606                         &$differenceEngine ] );
 607                 return $differenceEngine;
 608         }
 609
 610         /**
 611          * Get an appropriate SlotDiffRenderer for this content model.
 612          * @since 1.32
 613          * @param IContextSource $context
 614          * @return SlotDiffRenderer
 615          */
 616         final public function getSlotDiffRenderer( IContextSource $context ) {
 617                 $slotDiffRenderer = $this->getSlotDiffRendererInternal( $context );
 618                 if ( get_class( $slotDiffRenderer ) === TextSlotDiffRenderer::class ) {
 619                         //  To keep B/C, when SlotDiffRenderer is not overridden for a given content type
 620                         // but DifferenceEngine is, use that instead.
 621                         $differenceEngine = $this->createDifferenceEngine( $context );
 622                         if ( get_class( $differenceEngine ) !== DifferenceEngine::class ) {
 623                                 // TODO turn this into a deprecation warning in a later release
 624                                 LoggerFactory::getInstance( 'diff' )->info(
 625                                         'Falling back to DifferenceEngineSlotDiffRenderer', [
 626                                                 'modelID' => $this->getModelID(),
 627                                                 'DifferenceEngine' => get_class( $differenceEngine ),
 628                                         ] );
 629                                 $slotDiffRenderer = new DifferenceEngineSlotDiffRenderer( $differenceEngine );
 630                         }
 631                 }
 632                 Hooks::run( 'GetSlotDiffRenderer', [ $this, &$slotDiffRenderer, $context ] );
 633                 return $slotDiffRenderer;
 634         }
 635
 636         /**
 637          * Return the SlotDiffRenderer appropriate for this content handler.
 638          * @param IContextSource $context
 639          * @return SlotDiffRenderer
 640          */
 641         protected function getSlotDiffRendererInternal( IContextSource $context ) {
 642                 $contentLanguage = MediaWikiServices::getInstance()->getContentLanguage();
 643                 $statsdDataFactory = MediaWikiServices::getInstance()->getStatsdDataFactory();
 644                 $slotDiffRenderer = new TextSlotDiffRenderer();
 645                 $slotDiffRenderer->setStatsdDataFactory( $statsdDataFactory );
 646                 // XXX using the page language would be better, but it's unclear how that should be injected
 647                 $slotDiffRenderer->setLanguage( $contentLanguage );
 648                 $slotDiffRenderer->setWikiDiff2MovedParagraphDetectionCutoff(
 649                         $context->getConfig()->get( 'WikiDiff2MovedParagraphDetectionCutoff' )
 650                 );
 651
 652                 $engine = DifferenceEngine::getEngine();
 653                 if ( $engine === false ) {
 654                         $slotDiffRenderer->setEngine( TextSlotDiffRenderer::ENGINE_PHP );
 655                 } elseif ( $engine === 'wikidiff2' ) {
 656                         $slotDiffRenderer->setEngine( TextSlotDiffRenderer::ENGINE_WIKIDIFF2 );
 657                 } else {
 658                         $slotDiffRenderer->setEngine( TextSlotDiffRenderer::ENGINE_EXTERNAL, $engine );
 659                 }
 660
 661                 return $slotDiffRenderer;
 662         }
 663
 664         /**
 665          * Get the language in which the content of the given page is written.
 666          *
 667          * This default implementation just returns the content language (except for pages
 668          * in the MediaWiki namespace)
 669          *
 670          * Note that the page's language is not cacheable, since it may in some
 671          * cases depend on user settings.
 672          *
 673          * Also note that the page language may or may not depend on the actual content of the page,
 674          * that is, this method may load the content in order to determine the language.
 675          *
 676          * @since 1.21
 677          *
 678          * @param Title $title The page to determine the language for.
 679          * @param Content|null $content The page's content, if you have it handy, to avoid reloading it.
 680          *
 681          * @return Language The page's language
 682          */
 683         public function getPageLanguage( Title $title, Content $content = null ) {
 684                 global $wgLang;
 685                 $pageLang = MediaWikiServices::getInstance()->getContentLanguage();
 686
 687                 if ( $title->inNamespace( NS_MEDIAWIKI ) ) {
 688                         // Parse mediawiki messages with correct target language
 689                         list( /* $unused */, $lang ) = MessageCache::singleton()->figureMessage( $title->getText() );
 690                         $pageLang = Language::factory( $lang );
 691                 }
 692
 693                 Hooks::run( 'PageContentLanguage', [ $title, &$pageLang, $wgLang ] );
 694
 695                 return wfGetLangObj( $pageLang );
 696         }
 697
 698         /**
 699          * Get the language in which the content of this page is written when
 700          * viewed by user. Defaults to $this->getPageLanguage(), but if the user
 701          * specified a preferred variant, the variant will be used.
 702          *
 703          * This default implementation just returns $this->getPageLanguage( $title, $content ) unless
 704          * the user specified a preferred variant.
 705          *
 706          * Note that the pages view language is not cacheable, since it depends on user settings.
 707          *
 708          * Also note that the page language may or may not depend on the actual content of the page,
 709          * that is, this method may load the content in order to determine the language.
 710          *
 711          * @since 1.21
 712          *
 713          * @param Title $title The page to determine the language for.
 714          * @param Content|null $content The page's content, if you have it handy, to avoid reloading it.
 715          *
 716          * @return Language The page's language for viewing
 717          */
 718         public function getPageViewLanguage( Title $title, Content $content = null ) {
 719                 $pageLang = $this->getPageLanguage( $title, $content );
 720
 721                 if ( $title->getNamespace() !== NS_MEDIAWIKI ) {
 722                         // If the user chooses a variant, the content is actually
 723                         // in a language whose code is the variant code.
 724                         $variant = $pageLang->getPreferredVariant();
 725                         if ( $pageLang->getCode() !== $variant ) {
 726                                 $pageLang = Language::factory( $variant );
 727                         }
 728                 }
 729
 730                 return $pageLang;
 731         }
 732
 733         /**
 734          * Determines whether the content type handled by this ContentHandler
 735          * can be used for the main slot of the given page.
 736          *
 737          * This default implementation always returns true.
 738          * Subclasses may override this to restrict the use of this content model to specific locations,
 739          * typically based on the namespace or some other aspect of the title, such as a special suffix
 740          * (e.g. ".svg" for SVG content).
 741          *
 742          * @note this calls the ContentHandlerCanBeUsedOn hook which may be used to override which
 743          * content model can be used where.
 744          *
 745          * @see SlotRoleHandler::isAllowedModel
 746          *
 747          * @param Title $title The page's title.
 748          *
 749          * @return bool True if content of this kind can be used on the given page, false otherwise.
 750          */
 751         public function canBeUsedOn( Title $title ) {
 752                 $ok = true;
 753
 754                 Hooks::run( 'ContentModelCanBeUsedOn', [ $this->getModelID(), $title, &$ok ] );
 755
 756                 return $ok;
 757         }
 758
 759         /**
 760          * Returns the name of the diff engine to use.
 761          *
 762          * @since 1.21
 763          *
 764          * @return string
 765          */
 766         protected function getDiffEngineClass() {
 767                 return DifferenceEngine::class;
 768         }
 769
 770         /**
 771          * Attempts to merge differences between three versions. Returns a new
 772          * Content object for a clean merge and false for failure or a conflict.
 773          *
 774          * This default implementation always returns false.
 775          *
 776          * @since 1.21
 777          *
 778          * @param Content $oldContent The page's previous content.
 779          * @param Content $myContent One of the page's conflicting contents.
 780          * @param Content $yourContent One of the page's conflicting contents.
 781          *
 782          * @return Content|bool Always false.
 783          */
 784         public function merge3( Content $oldContent, Content $myContent, Content $yourContent ) {
 785                 return false;
 786         }
 787
 788         /**
 789          * Return type of change if one exists for the given edit.
 790          *
 791          * @since 1.31
 792          *
 793          * @param Content|null $oldContent The previous text of the page.
 794          * @param Content|null $newContent The submitted text of the page.
 795          * @param int $flags Bit mask: a bit mask of flags submitted for the edit.
 796          *
 797          * @return string|null String key representing type of change, or null.
 798          */
 799         private function getChangeType(
 800                 Content $oldContent = null,
 801                 Content $newContent = null,
 802                 $flags = 0
 803         ) {
 804                 $oldTarget = $oldContent !== null ? $oldContent->getRedirectTarget() : null;
 805                 $newTarget = $newContent !== null ? $newContent->getRedirectTarget() : null;
 806
 807                 // We check for the type of change in the given edit, and return string key accordingly
 808
 809                 // Blanking of a page
 810                 if ( $oldContent && $oldContent->getSize() > 0 &&
 811                         $newContent && $newContent->getSize() === 0
 812                 ) {
 813                         return 'blank';
 814                 }
 815
 816                 // Redirects
 817                 if ( $newTarget ) {
 818                         if ( !$oldTarget ) {
 819                                 // New redirect page (by creating new page or by changing content page)
 820                                 return 'new-redirect';
 821                         } elseif ( !$newTarget->equals( $oldTarget ) ||
 822                                 $oldTarget->getFragment() !== $newTarget->getFragment()
 823                         ) {
 824                                 // Redirect target changed
 825                                 return 'changed-redirect-target';
 826                         }
 827                 } elseif ( $oldTarget ) {
 828                         // Changing an existing redirect into a non-redirect
 829                         return 'removed-redirect';
 830                 }
 831
 832                 // New page created
 833                 if ( $flags & EDIT_NEW && $newContent ) {
 834                         if ( $newContent->getSize() === 0 ) {
 835                                 // New blank page
 836                                 return 'newblank';
 837                         } else {
 838                                 return 'newpage';
 839                         }
 840                 }
 841
 842                 // Removing more than 90% of the page
 843                 if ( $oldContent && $newContent && $oldContent->getSize() > 10 * $newContent->getSize() ) {
 844                         return 'replace';
 845                 }
 846
 847                 // Content model changed
 848                 if ( $oldContent && $newContent && $oldContent->getModel() !== $newContent->getModel() ) {
 849                         return 'contentmodelchange';
 850                 }
 851
 852                 return null;
 853         }
 854
 855         /**
 856          * Return an applicable auto-summary if one exists for the given edit.
 857          *
 858          * @since 1.21
 859          *
 860          * @param Content|null $oldContent The previous text of the page.
 861          * @param Content|null $newContent The submitted text of the page.
 862          * @param int $flags Bit mask: a bit mask of flags submitted for the edit.
 863          *
 864          * @return string An appropriate auto-summary, or an empty string.
 865          */
 866         public function getAutosummary(
 867                 Content $oldContent = null,
 868                 Content $newContent = null,
 869                 $flags = 0
 870         ) {
 871                 $changeType = $this->getChangeType( $oldContent, $newContent, $flags );
 872
 873                 // There's no applicable auto-summary for our case, so our auto-summary is empty.
 874                 if ( !$changeType ) {
 875                         return '';
 876                 }
 877
 878                 // Decide what kind of auto-summary is needed.
 879                 switch ( $changeType ) {
 880                         case 'new-redirect':
 881                                 $newTarget = $newContent->getRedirectTarget();
 882                                 $truncatedtext = $newContent->getTextForSummary(
 883                                         250
 884                                         - strlen( wfMessage( 'autoredircomment' )->inContentLanguage()->text() )
 885                                         - strlen( $newTarget->getFullText() )
 886                                 );
 887
 888                                 return wfMessage( 'autoredircomment', $newTarget->getFullText() )
 889                                         ->plaintextParams( $truncatedtext )->inContentLanguage()->text();
 890                         case 'changed-redirect-target':
 891                                 $oldTarget = $oldContent->getRedirectTarget();
 892                                 $newTarget = $newContent->getRedirectTarget();
 893
 894                                 $truncatedtext = $newContent->getTextForSummary(
 895                                         250
 896                                         - strlen( wfMessage( 'autosumm-changed-redirect-target' )
 897                                                 ->inContentLanguage()->text() )
 898                                         - strlen( $oldTarget->getFullText() )
 899                                         - strlen( $newTarget->getFullText() )
 900                                 );
 901
 902                                 return wfMessage( 'autosumm-changed-redirect-target',
 903                                                 $oldTarget->getFullText(),
 904                                                 $newTarget->getFullText() )
 905                                         ->rawParams( $truncatedtext )->inContentLanguage()->text();
 906                         case 'removed-redirect':
 907                                 $oldTarget = $oldContent->getRedirectTarget();
 908                                 $truncatedtext = $newContent->getTextForSummary(
 909                                         250
 910                                         - strlen( wfMessage( 'autosumm-removed-redirect' )
 911                                                 ->inContentLanguage()->text() )
 912                                         - strlen( $oldTarget->getFullText() ) );
 913
 914                                 return wfMessage( 'autosumm-removed-redirect', $oldTarget->getFullText() )
 915                                         ->rawParams( $truncatedtext )->inContentLanguage()->text();
 916                         case 'newpage':
 917                                 // If they're making a new article, give its text, truncated, in the summary.
 918                                 $truncatedtext = $newContent->getTextForSummary(
 919                                         200 - strlen( wfMessage( 'autosumm-new' )->inContentLanguage()->text() ) );
 920
 921                                 return wfMessage( 'autosumm-new' )->rawParams( $truncatedtext )
 922                                         ->inContentLanguage()->text();
 923                         case 'blank':
 924                                 return wfMessage( 'autosumm-blank' )->inContentLanguage()->text();
 925                         case 'replace':
 926                                 $truncatedtext = $newContent->getTextForSummary(
 927                                         200 - strlen( wfMessage( 'autosumm-replace' )->inContentLanguage()->text() ) );
 928
 929                                 return wfMessage( 'autosumm-replace' )->rawParams( $truncatedtext )
 930                                         ->inContentLanguage()->text();
 931                         case 'newblank':
 932                                 return wfMessage( 'autosumm-newblank' )->inContentLanguage()->text();
 933                         default:
 934                                 return '';
 935                 }
 936         }
 937
 938         /**
 939          * Return an applicable tag if one exists for the given edit or return null.
 940          *
 941          * @since 1.31
 942          *
 943          * @param Content|null $oldContent The previous text of the page.
 944          * @param Content|null $newContent The submitted text of the page.
 945          * @param int $flags Bit mask: a bit mask of flags submitted for the edit.
 946          *
 947          * @return string|null An appropriate tag, or null.
 948          */
 949         public function getChangeTag(
 950                 Content $oldContent = null,
 951                 Content $newContent = null,
 952                 $flags = 0
 953         ) {
 954                 $changeType = $this->getChangeType( $oldContent, $newContent, $flags );
 955
 956                 // There's no applicable tag for this change.
 957                 if ( !$changeType ) {
 958                         return null;
 959                 }
 960
 961                 // Core tags use the same keys as ones returned from $this->getChangeType()
 962                 // but prefixed with pseudo namespace 'mw-', so we add the prefix before checking
 963                 // if this type of change should be tagged
 964                 $tag = 'mw-' . $changeType;
 965
 966                 // Not all change types are tagged, so we check against the list of defined tags.
 967                 if ( in_array( $tag, ChangeTags::getSoftwareTags() ) ) {
 968                         return $tag;
 969                 }
 970
 971                 return null;
 972         }
 973
 974         /**
 975          * Auto-generates a deletion reason
 976          *
 977          * @since 1.21
 978          *
 979          * @param Title $title The page's title
 980          * @param bool &$hasHistory Whether the page has a history
 981          *
 982          * @return mixed String containing deletion reason or empty string, or
 983          *    boolean false if no revision occurred
 984          *
 985          * @todo &$hasHistory is extremely ugly, it's here because
 986          * WikiPage::getAutoDeleteReason() and Article::generateReason()
 987          * have it / want it.
 988          */
 989         public function getAutoDeleteReason( Title $title, &$hasHistory ) {
 990                 $dbr = wfGetDB( DB_REPLICA );
 991
 992                 // Get the last revision
 993                 $rev = Revision::newFromTitle( $title );
 994
 995                 if ( is_null( $rev ) ) {
 996                         return false;
 997                 }
 998
 999                 // Get the article's contents
1000                 $content = $rev->getContent();
1001                 $blank = false;
1002
1003                 // If the page is blank, use the text from the previous revision,
1004                 // which can only be blank if there's a move/import/protect dummy
1005                 // revision involved
1006                 if ( !$content || $content->isEmpty() ) {
1007                         $prev = $rev->getPrevious();
1008
1009                         if ( $prev ) {
1010                                 $rev = $prev;
1011                                 $content = $rev->getContent();
1012                                 $blank = true;
1013                         }
1014                 }
1015
1016                 $this->checkModelID( $rev->getContentModel() );
1017
1018                 // Find out if there was only one contributor
1019                 // Only scan the last 20 revisions
1020                 $revQuery = Revision::getQueryInfo();
1021                 $res = $dbr->select(
1022                         $revQuery['tables'],
1023                         [ 'rev_user_text' => $revQuery['fields']['rev_user_text'] ],
1024                         [
1025                                 'rev_page' => $title->getArticleID(),
1026                                 $dbr->bitAnd( 'rev_deleted', Revision::DELETED_USER ) . ' = 0'
1027                         ],
1028                         __METHOD__,
1029                         [ 'LIMIT' => 20 ],
1030                         $revQuery['joins']
1031                 );
1032
1033                 if ( $res === false ) {
1034                         // This page has no revisions, which is very weird
1035                         return false;
1036                 }
1037
1038                 $hasHistory = ( $res->numRows() > 1 );
1039                 $row = $dbr->fetchObject( $res );
1040
1041                 if ( $row ) { // $row is false if the only contributor is hidden
1042                         $onlyAuthor = $row->rev_user_text;
1043                         // Try to find a second contributor
1044                         foreach ( $res as $row ) {
1045                                 if ( $row->rev_user_text != $onlyAuthor ) { // T24999
1046                                         $onlyAuthor = false;
1047                                         break;
1048                                 }
1049                         }
1050                 } else {
1051                         $onlyAuthor = false;
1052                 }
1053
1054                 // Generate the summary with a '$1' placeholder
1055                 if ( $blank ) {
1056                         // The current revision is blank and the one before is also
1057                         // blank. It's just not our lucky day
1058                         $reason = wfMessage( 'exbeforeblank', '$1' )->inContentLanguage()->text();
1059                 } else {
1060                         if ( $onlyAuthor ) {
1061                                 $reason = wfMessage(
1062                                         'excontentauthor',
1063                                         '$1',
1064                                         $onlyAuthor
1065                                 )->inContentLanguage()->text();
1066                         } else {
1067                                 $reason = wfMessage( 'excontent', '$1' )->inContentLanguage()->text();
1068                         }
1069                 }
1070
1071                 if ( $reason == '-' ) {
1072                         // Allow these UI messages to be blanked out cleanly
1073                         return '';
1074                 }
1075
1076                 // Max content length = max comment length - length of the comment (excl. $1)
1077                 $text = $content ? $content->getTextForSummary( 255 - ( strlen( $reason ) - 2 ) ) : '';
1078
1079                 // Now replace the '$1' placeholder
1080                 $reason = str_replace( '$1', $text, $reason );
1081
1082                 return $reason;
1083         }
1084
1085         /**
1086          * Get the Content object that needs to be saved in order to undo all revisions
1087          * between $undo and $undoafter. Revisions must belong to the same page,
1088          * must exist and must not be deleted.
1089          *
1090          * @since 1.21
1091          * @since 1.32 accepts Content objects for all parameters instead of Revision objects.
1092          *  Passing Revision objects is deprecated.
1093          *
1094          * @param Revision|Content $current The current text
1095          * @param Revision|Content $undo The content of the revision to undo
1096          * @param Revision|Content $undoafter Must be from an earlier revision than $undo
1097          * @param bool $undoIsLatest Set true if $undo is from the current revision (since 1.32)
1098          *
1099          * @return mixed Content on success, false on failure
1100          */
1101         public function getUndoContent( $current, $undo, $undoafter, $undoIsLatest = false ) {
1102                 Assert::parameterType( Revision::class . '|' . Content::class, $current, '$current' );
1103                 if ( $current instanceof Content ) {
1104                         Assert::parameter( $undo instanceof Content, '$undo',
1105                                 'Must be Content when $current is Content' );
1106                         Assert::parameter( $undoafter instanceof Content, '$undoafter',
1107                                 'Must be Content when $current is Content' );
1108                         $cur_content = $current;
1109                         $undo_content = $undo;
1110                         $undoafter_content = $undoafter;
1111                 } else {
1112                         Assert::parameter( $undo instanceof Revision, '$undo',
1113                                 'Must be Revision when $current is Revision' );
1114                         Assert::parameter( $undoafter instanceof Revision, '$undoafter',
1115                                 'Must be Revision when $current is Revision' );
1116
1117                         $cur_content = $current->getContent();
1118
1119                         if ( empty( $cur_content ) ) {
1120                                 return false; // no page
1121                         }
1122
1123                         $undo_content = $undo->getContent();
1124                         $undoafter_content = $undoafter->getContent();
1125
1126                         if ( !$undo_content || !$undoafter_content ) {
1127                                 return false; // no content to undo
1128                         }
1129
1130                         $undoIsLatest = $current->getId() === $undo->getId();
1131                 }
1132
1133                 try {
1134                         $this->checkModelID( $cur_content->getModel() );
1135                         $this->checkModelID( $undo_content->getModel() );
1136                         if ( !$undoIsLatest ) {
1137                                 // If we are undoing the most recent revision,
1138                                 // its ok to revert content model changes. However
1139                                 // if we are undoing a revision in the middle, then
1140                                 // doing that will be confusing.
1141                                 $this->checkModelID( $undoafter_content->getModel() );
1142                         }
1143                 } catch ( MWException $e ) {
1144                         // If the revisions have different content models
1145                         // just return false
1146                         return false;
1147                 }
1148
1149                 if ( $cur_content->equals( $undo_content ) ) {
1150                         // No use doing a merge if it's just a straight revert.
1151                         return $undoafter_content;
1152                 }
1153
1154                 $undone_content = $this->merge3( $undo_content, $undoafter_content, $cur_content );
1155
1156                 return $undone_content;
1157         }
1158
1159         /**
1160          * Get parser options suitable for rendering and caching the article
1161          *
1162          * @deprecated since 1.32, use WikiPage::makeParserOptions() or
1163          *  ParserOptions::newCanonical() instead.
1164          * @param IContextSource|User|string $context One of the following:
1165          *        - IContextSource: Use the User and the Language of the provided
1166          *                                            context
1167          *        - User: Use the provided User object and $wgLang for the language,
1168          *                                            so use an IContextSource object if possible.
1169          *        - 'canonical': Canonical options (anonymous user with default
1170          *                                            preferences and content language).
1171          *
1172          * @throws MWException
1173          * @return ParserOptions
1174          */
1175         public function makeParserOptions( $context ) {
1176                 wfDeprecated( __METHOD__, '1.32' );
1177                 return ParserOptions::newCanonical( $context );
1178         }
1179
1180         /**
1181          * Returns true for content models that support caching using the
1182          * ParserCache mechanism. See WikiPage::shouldCheckParserCache().
1183          *
1184          * @since 1.21
1185          *
1186          * @return bool Always false.
1187          */
1188         public function isParserCacheSupported() {
1189                 return false;
1190         }
1191
1192         /**
1193          * Returns true if this content model supports sections.
1194          * This default implementation returns false.
1195          *
1196          * Content models that return true here should also implement
1197          * Content::getSection, Content::replaceSection, etc. to handle sections..
1198          *
1199          * @return bool Always false.
1200          */
1201         public function supportsSections() {
1202                 return false;
1203         }
1204
1205         /**
1206          * Returns true if this content model supports categories.
1207          * The default implementation returns true.
1208          *
1209          * @return bool Always true.
1210          */
1211         public function supportsCategories() {
1212                 return true;
1213         }
1214
1215         /**
1216          * Returns true if this content model supports redirects.
1217          * This default implementation returns false.
1218          *
1219          * Content models that return true here should also implement
1220          * ContentHandler::makeRedirectContent to return a Content object.
1221          *
1222          * @return bool Always false.
1223          */
1224         public function supportsRedirects() {
1225                 return false;
1226         }
1227
1228         /**
1229          * Return true if this content model supports direct editing, such as via EditPage.
1230          *
1231          * @return bool Default is false, and true for TextContent and it's derivatives.
1232          */
1233         public function supportsDirectEditing() {
1234                 return false;
1235         }
1236
1237         /**
1238          * Whether or not this content model supports direct editing via ApiEditPage
1239          *
1240          * @return bool Default is false, and true for TextContent and derivatives.
1241          */
1242         public function supportsDirectApiEditing() {
1243                 return $this->supportsDirectEditing();
1244         }
1245
1246         /**
1247          * Get fields definition for search index
1248          *
1249          * @todo Expose title, redirect, namespace, text, source_text, text_bytes
1250          *       field mappings here. (see T142670 and T143409)
1251          *
1252          * @param SearchEngine $engine
1253          * @return SearchIndexField[] List of fields this content handler can provide.
1254          * @since 1.28
1255          */
1256         public function getFieldsForSearchIndex( SearchEngine $engine ) {
1257                 $fields['category'] = $engine->makeSearchFieldMapping(
1258                         'category',
1259                         SearchIndexField::INDEX_TYPE_TEXT
1260                 );
1261                 $fields['category']->setFlag( SearchIndexField::FLAG_CASEFOLD );
1262
1263                 $fields['external_link'] = $engine->makeSearchFieldMapping(
1264                         'external_link',
1265                         SearchIndexField::INDEX_TYPE_KEYWORD
1266                 );
1267
1268                 $fields['outgoing_link'] = $engine->makeSearchFieldMapping(
1269                         'outgoing_link',
1270                         SearchIndexField::INDEX_TYPE_KEYWORD
1271                 );
1272
1273                 $fields['template'] = $engine->makeSearchFieldMapping(
1274                         'template',
1275                         SearchIndexField::INDEX_TYPE_KEYWORD
1276                 );
1277                 $fields['template']->setFlag( SearchIndexField::FLAG_CASEFOLD );
1278
1279                 $fields['content_model'] = $engine->makeSearchFieldMapping(
1280                         'content_model',
1281                         SearchIndexField::INDEX_TYPE_KEYWORD
1282                 );
1283
1284                 return $fields;
1285         }
1286
1287         /**
1288          * Add new field definition to array.
1289          * @param SearchIndexField[] &$fields
1290          * @param SearchEngine $engine
1291          * @param string $name
1292          * @param int $type
1293          * @return SearchIndexField[] new field defs
1294          * @since 1.28
1295          */
1296         protected function addSearchField( &$fields, SearchEngine $engine, $name, $type ) {
1297                 $fields[$name] = $engine->makeSearchFieldMapping( $name, $type );
1298                 return $fields;
1299         }
1300
1301         /**
1302          * Return fields to be indexed by search engine
1303          * as representation of this document.
1304          * Overriding class should call parent function or take care of calling
1305          * the SearchDataForIndex hook.
1306          * @param WikiPage $page Page to index
1307          * @param ParserOutput $output
1308          * @param SearchEngine $engine Search engine for which we are indexing
1309          * @return array Map of name=>value for fields
1310          * @since 1.28
1311          */
1312         public function getDataForSearchIndex(
1313                 WikiPage $page,
1314                 ParserOutput $output,
1315                 SearchEngine $engine
1316         ) {
1317                 $fieldData = [];
1318                 $content = $page->getContent();
1319
1320                 if ( $content ) {
1321                         $searchDataExtractor = new ParserOutputSearchDataExtractor();
1322
1323                         $fieldData['category'] = $searchDataExtractor->getCategories( $output );
1324                         $fieldData['external_link'] = $searchDataExtractor->getExternalLinks( $output );
1325                         $fieldData['outgoing_link'] = $searchDataExtractor->getOutgoingLinks( $output );
1326                         $fieldData['template'] = $searchDataExtractor->getTemplates( $output );
1327
1328                         $text = $content->getTextForSearchIndex();
1329
1330                         $fieldData['text'] = $text;
1331                         $fieldData['source_text'] = $text;
1332                         $fieldData['text_bytes'] = $content->getSize();
1333                         $fieldData['content_model'] = $content->getModel();
1334                 }
1335
1336                 Hooks::run( 'SearchDataForIndex', [ &$fieldData, $this, $page, $output, $engine ] );
1337                 return $fieldData;
1338         }
1339
1340         /**
1341          * Produce page output suitable for indexing.
1342          *
1343          * Specific content handlers may override it if they need different content handling.
1344          *
1345          * @param WikiPage $page
1346          * @param ParserCache|null $cache
1347          * @return ParserOutput
1348          */
1349         public function getParserOutputForIndexing( WikiPage $page, ParserCache $cache = null ) {
1350                 // TODO: MCR: ContentHandler should be called per slot, not for the whole page.
1351                 // See T190066.
1352                 $parserOptions = $page->makeParserOptions( 'canonical' );
1353                 if ( $cache ) {
1354                         $parserOutput = $cache->get( $page, $parserOptions );
1355                 }
1356
1357                 if ( empty( $parserOutput ) ) {
1358                         $renderer = MediaWikiServices::getInstance()->getRevisionRenderer();
1359                         $parserOutput =
1360                                 $renderer->getRenderedRevision(
1361                                         $page->getRevision()->getRevisionRecord(),
1362                                         $parserOptions
1363                                 )->getRevisionParserOutput();
1364                         if ( $cache ) {
1365                                 $cache->save( $parserOutput, $page, $parserOptions );
1366                         }
1367                 }
1368                 return $parserOutput;
1369         }
1370
1371         /**
1372          * Returns a list of DeferrableUpdate objects for recording information about the
1373          * given Content in some secondary data store.
1374          *
1375          * Application logic should not call this method directly. Instead, it should call
1376          * DerivedPageDataUpdater::getSecondaryDataUpdates().
1377          *
1378          * @note Implementations must not return a LinksUpdate instance. Instead, a LinksUpdate
1379          * is created by the calling code in DerivedPageDataUpdater, on the combined ParserOutput
1380          * of all slots, not for each slot individually. This is in contrast to the old
1381          * getSecondaryDataUpdates method defined by AbstractContent, which returned a LinksUpdate.
1382          *
1383          * @note Implementations should not call $content->getParserOutput, they should call
1384          * $slotOutput->getSlotRendering( $role, false ) instead if they need to access a ParserOutput
1385          * of $content. This allows existing ParserOutput objects to be re-used, while avoiding
1386          * creating a ParserOutput when none is needed.
1387          *
1388          * @param Title $title The title of the page to supply the updates for
1389          * @param Content $content The content to generate data updates for.
1390          * @param string $role The role (slot) in which the content is being used. Which updates
1391          *        are performed should generally not depend on the role the content has, but the
1392          *        DeferrableUpdates themselves may need to know the role, to track to which slot the
1393          *        data refers, and to avoid overwriting data of the same kind from another slot.
1394          * @param SlotRenderingProvider $slotOutput A provider that can be used to gain access to
1395          *        a ParserOutput of $content by calling $slotOutput->getSlotParserOutput( $role, false ).
1396          * @return DeferrableUpdate[] A list of DeferrableUpdate objects for putting information
1397          *        about this content object somewhere. The default implementation returns an empty
1398          *        array.
1399          * @since 1.32
1400          */
1401         public function getSecondaryDataUpdates(
1402                 Title $title,
1403                 Content $content,
1404                 $role,
1405                 SlotRenderingProvider $slotOutput
1406         ) {
1407                 return [];
1408         }
1409
1410         /**
1411          * Returns a list of DeferrableUpdate objects for removing information about content
1412          * in some secondary data store. This is used when a page is deleted, and also when
1413          * a slot is removed from a page.
1414          *
1415          * Application logic should not call this method directly. Instead, it should call
1416          * WikiPage::getSecondaryDataUpdates().
1417          *
1418          * @note Implementations must not return a LinksDeletionUpdate instance. Instead, a
1419          * LinksDeletionUpdate is created by the calling code in WikiPage.
1420          * This is in contrast to the old getDeletionUpdates method defined by AbstractContent,
1421          * which returned a LinksUpdate.
1422          *
1423          * @note Implementations should not rely on the page's current content, but rather the current
1424          * state of the secondary data store.
1425          *
1426          * @param Title $title The title of the page to supply the updates for
1427          * @param string $role The role (slot) in which the content is being used. Which updates
1428          *        are performed should generally not depend on the role the content has, but the
1429          *        DeferrableUpdates themselves may need to know the role, to track to which slot the
1430          *        data refers, and to avoid overwriting data of the same kind from another slot.
1431          *
1432          * @return DeferrableUpdate[] A list of DeferrableUpdate objects for putting information
1433          *        about this content object somewhere. The default implementation returns an empty
1434          *        array.
1435          *
1436          * @since 1.32
1437          */
1438         public function getDeletionUpdates( Title $title, $role ) {
1439                 return [];
1440         }
1441
1442 }